Ecological Economics 55 (2005) 203 – 217

www.elsevier.com/locate/ecolecon

ANALYSIS

Rationality, institutions and environmental policy

Arild Vatn*
Department of Economics and Resource Management, Norwegian University of Life Sciences, Norway
Received 2 July 2004; received in revised form 3 December 2004; accepted 6 December 2004
Available online 19 February 2005

Abstract

This paper is about how institutions determine choices and the importance of this for environmental policy. The model of
individual rational choice from neoclassical economics is compared with the model of socially determined behavior. While in
the first case, institutions are either exempted from or understood as mere economizing constraints on behavior, the latter
perspective views institutions as basic structures necessary also to enable people to act. The paper develops a way to integrate
the individualistic model into the wider perspective of social constructivism by viewing it as a special form of such construction.
On the basis of this synthesis three issues with relevance for environmental economics are discussed. First, the role of
institutional factors in the process of preference formation is emphasized. Next, the role of institutions for the choice of desired
states of the environment is analyzed. Finally, the effect of various policy instruments to motivate people to produce these states
is discussed. It is concluded that the core policy issue is to determine which institutional frameworks are most reasonable to
apply to which kind of problem. Issues, which from the perspective of neoclassical economics are pure technical, become
serious value questions if understood from an institutional perspective.
D 2005 Elsevier B.V. All rights reserved.

Keywords: Institutions; Rational choice; Social constructivism; Values; Preference formation; Policy formulation

1. Introduction It is intriguing to observe the different perspectives

on human choices prevailing in the literature. On the
Environmental choices are about the common one hand, we have the literature on economic man,
good. What interests me here is how we understand describing the human as a calculative, self-interested
and legitimize choices about this good. The issue is individual choosing under a set of external con-
thus both about how we come to define and value straints. On the other hand we have the concept of
environmental goods, and how we organize ourselves social man, seeing the human as socially constructed
to reach the goals we have set concerning their form with choices governed very much by internalized
and existence. values, rules and perceptions.
These two literatures are based on a very different
* Tel.: +47 6494 8606; fax: +47 6494 3012. perspective of rationality and the role of institutions.
E-mail address: [email protected]. The literature on economic man is methodological
0921-8009/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ecolecon.2004.12.001
204 A. Vatn / Ecological Economics 55 (2005) 203–217

individualist. Collective choice is seen as the sum of ality. I will briefly present the position, its limitations,
individual choices. Changes in behavior can only and the way institutional analysis has been attached to
come about by changing the constraints under which this model.
the individual acts—e.g., changes in relative prices or
punishment structures like the law. Legitimate 2.1.1. The economic model
changes in these constraints or institutions are, The core assumptions of neoclassical economics
furthermore, only those supported by individual are rational choice as maximization, stable prefer-
preferences. How to measure and aggregate these ences and equilibrium states (Becker, 1976; Eggerts-
has been a great challenge for this branch of literature son, 1990). Rationality is understood in individualistic
(e.g., Arrow, 1951; Gowdy, 2004). terms. It demands that preferences are complete,
The idea of social man is based on the methodo- transitive and continuous (Hausman, 1992). Further-
logy of social construction. Individuals are socialized more, to be termed rational, choices must be in
into patterns of thinking, roles and responsibilities. accordance with what is preferred the most by the
They act according to what is appropriate given the individual. As a logical consequence of the strong
actual situation. Thus, what become important issues version of methodological individualism2 that under-
is the result of a social process, and choices are more pins this model, preferences are treated as stable or at
seen as a reflection of internalized expectations. least as given.
Institutions are much more than constraints. They While economics in the above sense can be
are also what constitute the individual and create understood as a formal science studying the logic of
meaning. Within this tradition, the problem of a specific type of rational choice, its ambition is
legitimation is one about how well a specific also to say something about real world phenomena.
perspective or common solution fits the overall Then some assumptions about that world must be
picture of socially defined ideas. added. The standard application area for the
The aim of this paper is first to clarify more neoclassical core model is a world where agents
thoroughly how these two traditions view the relation- have full information, where transaction costs are
ship between the individual and the society. I will then zero, all resources are individually owned and
utilize these clarifications in reinterpreting the process competitive markets govern resource allocations
of formulating environmental policies—i.e., the proc- (Eggertsson, 1990).
ess from preference formation, via the definition of While there are institutional elements involved
which allocation of environmental resources is pre- here–i.e. property rights and markets–both the type of
ferred to the choice of policy instruments. The aim is rationality involved and the individual preferences are
thus to develop a better understanding of the assumed to be independent of the institutional context.
consequences of adapting either scheme to the whole The assumptions concerning full information and zero
field of environmental policy analyses. While such an transaction costs3 imply that there are no problems
aspiration certainly reduces the possibility to go into connected to the perception of alternatives and the
much detail, it offers the opportunity to emphasize communication of needs via these institutions. While
important links across a significant field.1 the model is about coordination–i.e., exchange of
goods–the coordination process itself is understood as
immediate and costless. The paradigm is furthermore
2. Choice, rationality and institutions

2.1. The individualist perspective 2

By strong is implied that all social phenomena are in principle
explicable only in terms of independent individual acts. A weaker
The neoclassical economic model is considered to version of methodological individualism accepts that external forces
capture the essence of individual, calculative ration- may change the individual, but insists that studying choice still has
to be based on studying the actions of these individuals. Elster
(1983) is a representative of this latter position.
1 3
A more in depth analysis of these issues is found in Vatn Defined as costs of information gathering, contracting and
(2005). controlling (Dahlman, 1979).
 A. Vatn / Ecological Economics 55 (2005) 203–217 205

based on a positivist/empiricist perception of the economists that base their analysis on the neoclassical
world. Knowledge is a result of instantaneous core, but dispense from the assumptions of full
experience and perception. information and zero transaction costs.4
Given the standard core and application area, this Positive information costs imply that uncertainty is
literature offers an internally consistent structure. introduced into the model. There may be uncertainty
Problems may, however, occur when we use the about the quality and quantity of a good. Behavior
model to study existing economies. I will focus on becomes costly to observe, thus opportunism may
two issues; the problem of positive information costs prevail. Observing the different institutional structures
and the costs of attaching property rights to goods/ of modern market economies, authors like North
benefit streams—i.e., positive demarcation (trans- (1990), Eggertsson (1990), Furubotn and Richter
action) costs. (1998), and Williamson (1985, 2000)5 emphasize that
Concerning the costs of information, the issue of these structures are introduced to reduce uncertainty
information asymmetries has been taken on board and economize on transaction costs.
(e.g., Akerlof, 1970). While yielding interesting These authors typically define institutions as
results, dispensing from the assumption of full constraints on individual behavior. North (1990:3),
information and still assuming maximizing behavior sees institution as bthe rules of the game in a society
can be shown to be inconsistent. We will return to this or, more formally, . . . the humanly devised constraints
later. that shape human interaction.Q To the degree that
Turning to the costs of demarcating benefit rational choice and equilibrium are continued as core
streams, these may be high and make it impossible assumptions of the model, the vision is that institu-
to exclude anybody from the consumption of a tions appear as constraints on a standard optimization
specific good. We enter the arena of public goods problem.
and bads. In such situations markets cannot allocate Still, accepting positive information/transaction
resources in a sensible way, and according to neo- costs challenges the whole idea of optimization. To
classical welfare theory it is necessary to invoke some define optimum precisely, information has to be
type of institutional structure other than the market. complete—i.e., information costs have to be zero. If
One solution to this problem has been to introduce the information gathering is costly, the actor will have to
social planner. This answer has not gotten unanimous decide whether resources should be used on conduct-
consent among economists, not least because it ing choices or on gathering more information as a
implies dispensing with the methodological individu- basis for a choice that could be even better. The
alist perspective itself. It has also been attacked for its character of information–i.e., the fact that you can
inconsistent treatment of transaction costs—e.g. never know what is the value of knowing more–
Coase (1960). If transaction costs are zero, as makes it impossible to determine an optimal solution
assumed, there should be no problem for markets to to this problem (Knudsen, 1993). Thus it seems
secure optimal amounts of public goods/bads. Demar- inconsistent to continue keeping the assumption of
cation and exclusion would then be cost free. Given individual rationality as defined intact when adding
the individualist perspective, zero transaction costs
really imply no public goods. If they are positive, lack
of supply may just be an efficient reaction to such 4
There is some confusion in the literature concerning the naming
costs.
of various positions. Eggertsson (1990), as an example, distin-
guishes between dnewT (e.g., Williamson, 1985) and dneoT (e.g.,
2.1.2. dNeoclassicalT or dnewT institutionalism North, 1990, Eggertsson, 1990 himself) institutionalism. He
The above issues–the question of information/ emphasizes that it is only the dneoT institutionalists that are dtrueT
transaction costs and the idea of the social planner– to the neoclassical core. Williamson accepts dsatisfyingT or
dbounded rationalityT—see also note 5. Here I use dnewT to cover
have institutional implications that have been dealt
both positions.
with extensively in close association with neoclassical 5
It should be emphasized that Williamson dispenses from the
economics itself. The position of dnewT institutional standard definition of the neoclassical core by also accepting
economics will here be used to describe institutional bounded rationality.
206 A. Vatn / Ecological Economics 55 (2005) 203–217

the assumption of positive information and transaction are many different positions in this literature. While
costs to the model. the virtue of the neoclassical model is its highly
Turning to the issue of social welfare and the role developed and rather unified formalism, a less
of the social planner, we observe yet another set of unifying structure can be found in the various
consistency problems. In welfare theory the rule for traditions of social constructivism. Still, some com-
the planner is to maximize social welfare, mainly by mon themes are very visible.
creating solutions as if markets exist even for public The origin of these perspectives can be traced as
goods. Thus the same type of institutional frame and far back as to Rousseau (1762/1968), later to the
calculative rationality dominates both market and sociologists Emile Durkheim (e.g., Durkheim, 1893/
non-market allocations. It is consumers’ individual 1949, 1895/1950) and the founder of dclassicalT
willingness to pay that is the criterion in both institutional economics Thorstein Veblen (e.g.,
situations. On the other hand, the planner is devised Veblen, 1898, 1899). Within modern sociology we
to represent the will of the consumers. He is thus may separate two main traditions, the cognitive and
instituted as a dself-sacrificingT agent for the common the normative. A similar structure is found within
will as expressed by the fictitious market. This implies dclassicalT institutional economics—both among the
actually to import the idea of role defined behavior doldT (e.g. Veblen, 1898, 1899) and the dmodernT7
into the model, a concept that is very far from the (e.g., Bromley, 1989; Hodgson, 1998; Tool, 1995).
methodological individualist perspective dominant in Here the distinction is, however, made between three
economics. There is rarely any explicit discussion in different aspects of institutions. They are partly seen
the literature concerning the consistency of establish- as constraints, as cognitive models, and finally as
ing such a capacity and on ways to institutionalize it, normative structures (Dequech, 2002). While the
however. perspective of institutions as constraints is already
The idea of the benevolent planner, dispensing covered, I will in the following emphasize the core
with individual rationality, is naturally attacked, not messages concerning the cognitive and the normative
least in the public choice brand of economics itself. aspects. I have chosen to base my presentation on the
Here institutionally defined choice structures are two perspectives as they appear in sociology.
completely neglected. There is only selfishness. The
model of calculative market behavior is thus trans- 2.2.1. The cognitive aspect
ferred also to the domain of policy making itself. The work of Berger and Luckman (1967) is basic
Planning becomes mere dpoliticsT—a market for the to the cognitive position. In their understanding, the
realization of the interests of policy makers or basic problem is how people can communicate and
administrators themselves (Buchanan, 1978; Dear- cooperate in a complex world, a world that cannot
love, 1989; Niskanen, 1971). According to this line of explain itself to us. Common knowledge in the form
thinking, the problem is the lacking ability of people of concepts or typifications of both natural and social
to make fair or efficient decisions on behalf of others. phenomena constitutes the basis for creating neces-
It is considered to be against human nature. sary meaning and order so that cooperation becomes
possible. In this understanding the world as we see it
2.2. The social constructivist perspective on is socially constructed. It is perceived through the
institutions concepts we collectively produce.
Their understanding is thus very different from the
In sociology and dclassicalT institutional econom- individualist where it is taken for granted that
ics6 behavior is dominantly understood as socially individuals have a predefined ability to understand
created, implying that choices reflect norms, rules, both own needs, others’ performance and the working
and expectations built into the institutions of a society.
The social dimension is an objective reality, having 7
The dmodern classicalT tradition is often named neoinstitutional
distinct effects on the shaping of the individual. There economics (e.g., Tool, 1995). Since this concept is also used to
name the most important position within dnewT institutional (see
6
That is the tradition from Veblen and Commons. footnote 3), I will avoid it here.
 A. Vatn / Ecological Economics 55 (2005) 203–217 207

of the natural world. In the social constructivist ations. Roles are not only structuring life through
perspective not only this understanding itself, but creating reciprocal expectation. The value of doing the
also the fundamental cognitive structures by which right thing is emphasized.
any issue has to be treated, must be created. The core March and Olsen (1989) represents a recent
example is the development of the language. The version of this position. They emphasize that
development of the cognitive takes the form of a bbehavior is contained or dictated by cultural dicta
dialectical process. Society, as an objective reality, is and social norms. Action is often based more on
both a product of human action and creates its identifying the normatively appropriate behavior than
members. on calculating the return expected from alternative
There are three phases in this dialectic process: choices.Q (March and Olsen, 1989: 22). Rationality is
First we have externalization, implying that sub- to do what is appropriate. Still, they emphasize that
jectively constructed typifications like a dmealT or a behavior is not unreasoned. Every situation must be
dmeasurement scaleT are expressed; next we have interpreted and one must always choose what one
objectivation where these typifications retain exis- finds to be the appropriate response. The point is that
tence independent of those creating them and stand the role and the situation define very much what is
out as facts or dthingsT; finally there is internal- acceptable.
ization, the phase in which these social constructs The distinctions made between the cognitive and
are taken up and reproduced by others. Through the normative positions are mainly those of degree.
objectivation the social constructs take on a form of Both focus on how conventions or norms help sorting
dnaturalityT. They become given and hard to observe out complex choice situations. While the cognitive
as socially constructed. For those growing up, they perspective focuses on understanding what is at stake
are just there. and classifying relevant behavior, the normative
According to Berger and Luckmann it is the perspective is more oriented towards the prescriptive
sharing of the classification, the common acceptance domain of institutions. They are developed to handle
of a typification, which makes it an institution. They situations where conflicts may appear or certain
further stress that bthe origins of any institutional values are considered important to reproduce. The
order lie in the typification of one’s own and other’s cognitive focus on conventions is more neutral and
performances. This implies that one shares with others more basic. It focuses on shared definitions of social
specific goals and interlocking phases of performance, reality. There is, however, nothing in the cognitivist
and, further, that not only specific actions but forms of position that precludes norms of the kind described
action are typifiedQ (op. cit.: 89). These typifications here.
are objectified not least through roles.
This is a very different understanding also of 2.3. Towards a synthesis
institutions as compared to that of dnewT institutional
economics. Institutions are what enable humans to The two main perspectives addressed here, the
act. This is an element also strongly emphasized by individualist and the social constructivist, are pro-
dclassicalT institutional economists. Furthermore, they foundly different. To produce a synthesis could thus
are important in the creation of the individual and give be considered infertile. Still, I think it is both possible
necessary meaning to the actions s/he undertakes. and productive to do so. The basis for a synthesis is
Social identity is produced thereby (Scott, 1995). the cognitivist assertion that we construct the way we
see the world. Through the development of the
2.2.2. The normative aspect language and various cognitive models, it becomes
The normative perspective or tradition shares the possible to handle the complex situations we are
idea of symbolic interaction with the cognitivist confronted with and produce necessary predictability
position. To the extent that there is a difference, the concerning action in the social sphere. The typifica-
normative tradition has a stronger emphasis on the tions involved are necessary for us to make sense of
creation of common values and the pressure placed on sense data. They play, however, also a direct role in
individuals to fulfill certain obligations and expect- coordinating behavior. They define alternatives and
208 A. Vatn / Ecological Economics 55 (2005) 203–217

construct standards, thus bringing complexity down to certain situation. Just by assuming others to behave
a manageable level.8 calculatively rational, we can make sense of what
Given this basis, institutional structures may be other persons are doing (e.g., Elster, 1990). Still,
developed that support different types of rationalities. assuming something does not make it real. Instead, we
They might support individual calculative behavior or may turn the idea around and look at some institutions
they might support more normative behavior—i.e., as developed to foster calculative rationality. Humans
they focus on dIT or some concept of a dWeT. While the have over the years constructed a lot of devices
model of individual rational choice defines rationality making calculation possible—e.g., the concept of
as a universal structure, the perspective taken here is money and the double entry book-keeping, which
that the logic of the situation is institutionally formed. both are instituted to foster calculation and help
There is no single form of rationality, which is people compensate for deficiencies in their cognitive
institutionally independent. Maximizing individual capacities. The market itself is maybe the example par
utility may fit one set of institutional structure, but excellence of how humans have created institutions to
be impossible, irrelevant or embarrassing in other foster calculation. Referring to the variety across
situations. cultures concerning exchange, Bowles (1998:90)
The model of individual maximization focuses on emphasizes that b[m]arkets are thus powerful cogni-
independent calculation. It is, however, not possible, tive simplifiers, allowing radical reductions in the
even in context supporting individual rationality, to complexity with which one typically views an assort-
dispense from the fact that no cognitive structure is ment of disparate goods.Q
readily available to us as unsocialized individuals. While methodological individualists claim that the
Furthermore, concepts cannot be understood inde- process of choosing can be strictly demarcated from
pendent of the context. Thus, the basic cognitive the process of developing ones goals or preferences,
structures we utilize, whether we maximize or not, the cognitivist paradigm implies that the typification
have to be socially constructed and reproduced. itself influences both these levels. They are inter-
Whether nature has given humans the capacities to linked. The culturally dependent typifications are also
stand up to the dnormT of maximizing is then yet important due to their capacity to produce meaning.
another issue. Maybe even this ability has at least to These meanings are internalized via various pro-
be institutionally supported to work—i.e., bounded cesses. Certainly, there are individual tastes involved.
rationality. Still, the fairly low variations in preferences within a
From this it follows that the individual, even in cultural group compared with the great heterogeneity
contexts where it is appropriate to calculate individual across groups, supports the idea that preferences are
gains, is dependent on the support of cognitive also social constructs to a large extent. As such they
structures, which are collectively produced. These can be evaluated, debated and disputed in the
structures make it possible to process the sense data community of individuals—among the members of a
that is needed for doing the relevant calculations in dWeT. One may argue that certain types of preferences
such a setting. This basis is always needed for are preferable to hold either for the individual, the
communication and coordination among individuals. society or for both. This is in itself a process that is
The process of internalization, and the fact that all part of the typification and lends itself to normative
members of a community use the same constructs, interpretations.
may make this aspect invisible to us. Thus, no perspective on choice can escape the fact
Some claim that rational choice itself solves the that the concepts used, the typifications invoked, are
problem of understanding what is going on in a socially constructed. Neoclassical economics is not a
natural given. It is a special type of social construct—
8
assuming individuality, calculation and costless inter-
An illustration of this can be given from modern brain research action. It is one of many typifications structured in
indicating that while the amount of sense data a body receives per
second is about 11 million bits, our consciousness is only able to clusters of reciprocal roles that constitute institutional
handle 10–30 bits dependent on training—e.g., predefined typifi- arenas of different kinds of rationalities like the
cations (Nbrretranders, 1991). market, the family, the polity, etc. Each of these
 A. Vatn / Ecological Economics 55 (2005) 203–217 209

systems is constructed to support specific types of allocation of these goods. Finally, the effects of
values and meanings. Thus some structures, such as various policy instruments used to influence behavior
the market, support calculative rationality and com- so as to obtain this allocation will be discussed. The
petition. Others, like the family, support care. Norms most important question in the latter case is whether
and conventions are, however, part of any such such instruments can be viewed as neutral, or if they
structure. influence the choice situation also by affecting the
Furthermore, given positive information/transac- motivations and rationalities that get dactivatedT.
tion costs, calculative reasoning may not be enough
for the market to work properly. Norms of dgood 3.1. Preference formation
conductT are invoked. These norms function as a
guard against opportunism and fraud, which is itself a In the last 10–15 years we have experienced an
logical consequence of individual rationality in the intense debate over the quality of research undertaken
context of positive transaction costs. If you can get to measure the value of various environmental
away with cheating someone, the individualist logic amenities in monetary terms. This debate has not
tells you that you should (Kapp, 1971). A dominant least concerned which type of preferences people have
aspect of institutionalization in the marketplace seems for such services. The debate was fuelled by a large
to be to reduce transaction costs, to make transactions amount of observations where individualsT responses
easier, to include more transactions into the market, deviated from the assumptions of the economic model
and to increase the chance to be caught if acting (Vatn, 2004). Diamond and Hausman (1994) were
opportunistically. Some of these aspects are rather among those defending this model by concluding that
instrumental. Others are, however, more normative. people do not have preferences over environmental
Thus normative aspects are an important part also of goods. They simply lack bexperience in markets for
the market structure. This goes for any institutional environmental commoditiesQ as they termed it (op.
structure. Still, which types of norms and which role cit.: 62). While invoking the economic model as the
they play in the choice situations will certainly vary. frame of reference, they concluded that environmental
The above have important implications for envi- valuation does not fall within this domain.
ronmental economics. The problem with the standard Gregory et al. (1993) observed the same type of
position is that it involves the logic of markets without problem. Asserting that people have well-defined
asking whether this is the perspective which is values only for familiar goods, they proposed a
appropriate when analyzing common goods. It may constructivist procedure to support respondents when
be a category mistake to invoke the market perspec- valuing environmental goods. They found it necessary
tive here, since the issue is really about dWeT not about to use a kind of an inferential process where the
dIT. If one believes, however, that individual, calcu- individual was supported by the analyst when devel-
lative rationality is all there is or at least the most oping or uncovering her preferences. Hanemann
developed form of reason, one will not recognize this (1994) did not find this to be problematic for the
problem. Instead importing the logic of individual economic model. He emphasized that b[t]he real issue
maximization, as in the market, will stand out as is not whether preferences are a construct but whether
bringing rationality into yet another arena. The next they are a stable constructQ (p. 28). The question then
step will be to sort these things out. just becomes helping the individual to learn about her/
himself.9
What we observe here is how a conceptualization
3. Consequences for environmental policy analyses based on the economic model is used to typify and
discuss issues that, in this case, do not fit too well.
When applying the above understanding to the
field of the environment, I will divide the analysis into 9
It should be noted that it may also be a problem for the
three parts. First, I will look at the formation of consistency of the model if learning about ones own preferences is
preferences about environmental goods. Next, I will costly. Again this complicates the definition of an optimum (confer
focus on the process of making choices over the Section 2.1). This is still not the issue here.
210 A. Vatn / Ecological Economics 55 (2005) 203–217

Thus the explanations themselves can be seen to environment is a field where normative issues are
emanate from the underlying social construction. We likely to frequently appear. This has two important
observe that the specific type of dconstructivismT implications. First, people may want to take part in
proposed by Gregory et al. is directed towards discussions over which values to hold rather than just
processes that are internal to the individual. The great expressing individual positions. I will return to this in
challenges for neoclassical economics appear first if Section 3.2. Second, one should expect that the
preferences are seen also to be socially contingent. normative aspects involved would result in individual
This issue is, however, not addressed within the preferences with restricted trade-off abilities. The
environmental valuation literature itself. latter conclusion is based on the logic that normative
From a constructivist perspective the social plays a rules or principles are not (easily) traded against
central role in forming preferences. Certainly, needs ordinary commodities—e.g., monetary values
like food and shelter are physically determined. Still, (Etzioni, 1988). People may view these concerns
the way even such needs are fulfilled is socially rather as citizen issues (right and wrong) than ordinary
contingent. According to the same perspective, it is no consumer questions (exchange value). If such princi-
surprise that people may not have preferences for ples are evoked, consumer behavior may be influ-
many environmental issues. They are new to us, a enced by citizen norms, as in the case of boycotts
consequence of the ever-changing production pro- against products produced in e.g., environmentally
cesses of modern societies. The problem is, however, unfriendly ways.
not just lack of experience. Taking stand demands Restricted trade-offs may appear in the form of
typifications—i.e., concepts and models. These will lexicographic preferences, frequently observed in the
also be lacking if the social process of shaping environmental valuation literature, confer Stevens et
cognition is lagging behind, as typically happens for al. (1991), Spash and Hanley (1995), Spash (2000).
environmental issues in a world characterized by rapid Yet, following our perspective, what is an issue for the
changes. Not only may the issues be unfamiliar. individual or the community, what is a citizen and
Standardized responses may yet not be developed. what is a consumer issue, is itself influenced by the
In relation to this it must be mentioned that there is way we construct and envision the problems. This
a renewed10 interest in viewing preferences as moves us from the level of preferences to that of value
endogenous to the economic system—e.g., Bowles articulating institutions.
(1998), Hodgson (1998), Lane (1991), and Sunstein
(1993). Bowles (1998) discusses a wide range of 3.2. The process of choosing desired states of the
relationships between preferences and the structure of environment
economic institutions. While he emphasizes the
importance of institutions in forming individual If context does not matter for the preferences we
preferences, norms and behavior, he also accentuates hold, the frames invoked when eliciting values should
that we know very little about the mechanisms that not be important either. If we instead base our
explain these relationships. understanding on the synthesis from Section 2.3, the
Certainly, in some areas of life socialization and institutional setting may itself have implications for
norms become especially important. This is not least which preferences we express.
the case for issues where our preferences influence the The core concept here is that of value articulating
possibilities of others. Environmental issues are a institutions (Jacobs, 1997). The type of institutions
typical example of this category, in particular because invoked in the process of expressing values, influen-
of its common good characteristics. Thus, the physical ces both the values that come forward and the
conclusions that can be drawn on the basis of them.
10
The effect of institutions on desires or preferences was Using real or imitated markets (contingent valuation
certainly an issue that was much acknowledged by the classical (CV)) and cost–benefit analysis (CBA) will make
economists like Adam Smith, John Stuart Mill and Karl Marx.
Certainly, in the brand of dclassicalT institutionalists, this is a basic respondents focus on monetizing and trade-offs.
insight as developed from Veblen (confer Groenewegen et al., Using multicriteria analysis implies (a) accepting
1995). values to be expressed in different, even incommen-
 A. Vatn / Ecological Economics 55 (2005) 203–217 211

surable, dimensions and (b) invoking a more commu- The authors conclude in the following way: bWhen
nicative process (Munda, 2004). Using participatory deconstructed by the respondents themselves, their
processes like focus groups, consensus conferences or WTP figures proved to have little substance and they
citizens’ juries, offers further emphasis to the com- unequivocally rejected CV as an acceptable means
municative aspect. of representing their values . . . valuing nature in
A value articulating institution is a constructed set monetary terms was incommensurable with deeply
of rules or typifications. It defines who shall held cultural valuesQ (Clark et al., 2000: 60). Other
participate and on the basis of which capacity—i.e., studies, like Vandjal and O’Connor (1994) and
in which role. Thus in a CV study we are supposed to Schkade and Payne (1993), give similar results.
respond as consumers. In a citizens’ jury we are An especially strong feature of the Clark et al. study
supposed to act on behalf of a society—as citizens—a was the observation that people, while being inter-
role, which is supported by the structure of the viewed, did not really understand for what reason the
institution. willingness to pay figures were collected. The context
A value articulating institution also defines what is was not at all obvious to them. When they afterwards
considered relevant data and how data is to be understood that the data were to be used in a CBA
handled. In the case of CV, monetary bids are the framework to support environmental decisions, in this
relevant facts. In the case of citizens’ juries, it is the case nature conservation, many even felt deceived.
argument that counts. In the case of CV/CBA data are The idea that respondents are searching for the
aggregated uniformly across all respondents and costs dlogic of the situationT may help to explain yet another
(gains) at different times are compared through anomaly—i.e., why some people having lexico-
discounting. In the case of a citizen jury the graphic preferences nevertheless pay in CV studies
importance of the argument, its fit really to the (Spash, 2000). The CV institution demands a mone-
cognitive framework invoked, is what is decisive. tary expression. People thus respond within the
The main point here is that using different value context they are asked to behave. Protest bids show,
articulating institutions will tend to give different however, that at least some do still not obey the
outcomes or preferred solutions. Thus the choice of framework.
such institutions is certainly non-trivial. Furthermore, If values of different categories are involved, it will
the people involved in the process will most probably be inconsistent to use a value articulating institution
have ideas about which frames they find most based on commensurability—e.g., monetary represen-
relevant. One should even expect that they would tation (O’Neill, 1993). Incommensurable values must
react negatively to frames that are not found relevant be treated by other rationalities than the calculative
or acceptable—e.g., using individual-oriented value type of reasoning. Environmental issues are about
articulating institutions for something considered ethical, not so much about individual values (Holland,
collective and vice versa. 1995). As such they are mainly about conflict
dProtest bidsT in the form of refusals to pay is a resolution where communicative rationality (Haber-
commonly observed feature of CV studies. Still, few mas, 1984) instead of a calculative one offers a more
studies have focused on why people act this way. reasonable logic.
Spash and Hanley (1995) document results from such As emphasized by Sunstein (1993), if preferences
an analysis, a CV study focusing on a lowland heath are affected by the institutional context, one cannot
in Britain. In this case many people reacted against the base (environmental) regulations simply on an aggre-
dcommodificationT that was implicit in the payment gate of private preferences. This would produces mere
vehicle. Heaths should be common land. In other circularities. While we cannot resolve such dilemmas,
studies—e.g., Clark et al. (2000), and Spash (2000)— communication about which contexts we find to be
ethical concerns like the right of certain species to the most relevant for expressing environmental values
exist also motivate protest bids or at least ambivalence is an important second-order issue. Such processes
towards using payment as a measure. may stimulate the creation of typifications or rules
The study by Clark et al. (2000) is interesting also concerning which institutions have merit in which
because it focuses on how people construct CV bids. situations. This is the way societies have treated these
212 A. Vatn / Ecological Economics 55 (2005) 203–217

issues all along (Douglas, 1986; Walzer, 1983). Only punish if met with hostility, even if this would be
in modern societies did the idea emerge that markets costly for them.
can be a universal mechanism for determining all In the case of public goods games, the standard
types of resource use problems. Sunstein (1993: 252) solution following from the assumption of strategic
concludes bthat much of public discussion is about the behavior is that defecting is the rational thing to do. It
appropriate mode (not only the appropriate level) of is, however, argued that if games are viewed as
valuation for diverse goodsQ (author’s emphasis). continuous, instead of a finite number of rounds, the
These are altogether strong arguments for putting so-called Folk theorem can be maintained (Fudenberg
much more effort into developing methods that better and Tirole, 1991; Romp, 1997). It concludes that if the
take into account the character of the goods and the game is repeated infinitely and people are sufficiently
kind of values involved—e.g., participatory value patient–i.e., future gains from cooperation are not
articulating institutions. discounted dtoo muchT–it becomes individually
rational to cooperate.
3.3. Policy instruments and the effect of invoking Studies like Fehr and Gächter (2000) and Ostrom
different logics (2000) show, however, that participants are willing to
cooperate even if the game is not replicated, or if the
If the institutional context influences which pref- composition of the group changes if replicated, etc.
erences are formed or evoked, they may also affect the They also show that cooperation takes place even in
kind of behavior that is stimulated. Thus the choice of the last round of a sequence of finite games. Fehr and
policy instruments may influence not only the pay- Gächter (2000) argue that people act reciprocally
offs for the agents involved, as when taxes are used, instead of strategically.
but also the logic or rationality agents come to invoke. Studies covering typical principal-agent relations
The literature on these issues is developing. While give similar results as they also illustrate the
we must wait for further research to draw more importance of trust. Here the formulation of incentive
specific conclusions, a rather consistent pattern seems or control structures is the core issue. In a study of
to evolve explaining a range of observations deviating sanctioning systems including fines for environmen-
from the predictions of the rational choice paradigm. tally bad conduct, Tenbrunsel and Messick (1999)
It is shown that reciprocity and trust may be as observe that control by the principal may actually
important as individual rationality or strategic behav- reduce compliance by the agents up to a certain level
ior in explaining outcomes. Parallel, normative of punishment before it starts increasing. Similar
motivations may play crucial roles. observations are done in other studies–e.g., Gneezy
The literature on so-called dultimatum gamesT11 and Rustichini (2000). This literature indicates that
(Güth et al., 1982; Gintis, 2000) and dpublic goods control may reduce compliance because it signals that
gamesT12 (Ostrom, 2000) shows the importance of the principal does not have trust in the agent. The
reciprocity. People are willing to cooperate or share intrinsic motivation for compliance is actually
under the presumption that others do the same. They reduced. Beyond a certain level of punishment, the
act friendly when met by friendly actions. They negative effect of the control becomes so high that the
agent still must take account of it. Indeed, control of
11
In these games, the so-called proposer gets a sum of money– those acting individually strategic (the dITs)–i.e.,
e.g., 10 dollars–which s/he must split between her-/himself and a notoriously breaking the law or agreements–tends
respondent. The players do not know each other, and the game is
not repeated. If the respondent accepts the division, both players get
not to have this dunexpectedT effect. The fact that the
the money on the basis of the split made by the proposer. If the principal reacts towards these agents seems to
respondent turns the offer down, the two participants do however reinforce the positive self-image of those complying
get nothing. freely (the dWeTs) (Fehr and Falk, 2002).
12
This is a type of experiments where a multiple prisoners’ In recent research on dincentive wagesT–i.e., wages
dilemma game is created—i.e., a structure where the best results are
obtained if everybody cooperate, while it is beneficial for the
that depend on monitored effort–the existence of
individual to defect. The results are especially beneficial for the dreciprocity-driven voluntary cooperationT is consis-
individual if all the others cooperate. tently observed (Fehr and Gächter, 2002). This study
 A. Vatn / Ecological Economics 55 (2005) 203–217 213

The rationality of the situation as perceived by agents

I We
A: B:
I Strategic ‘Crowding
The rationality of the situation behavior out’
as perceived by the principal
C: D:
We ‘Crowding Reciprocity/
in’ obligation

Fig. 1. Perceptions, policy measures and behavior.

itself suggests that positive incentives for increased Frey (1997b) shows how different constitutional
effort might even result in decreased effort compared arrangements affect predisposition to tax avoidance.
to fixed wages defined against some expected effort. The above literatures tend to rest mainly on
They also find that the reference point in incentive psychological explanations. Crowding out is thus
programs matter. Despite the fact that the payment typically explained by so-called intrinsic motivation.
structure is identical, incentives that dpunishT low While it seems to be driven by other motivations than
instead of drewardingT high production yield much the external reward, I believe it is more relevant to
lower effort. Fehr et al. (1997) observe that monitor- explain what we see by thinking in terms of
ing and the use of fines even in cases of verified internalized (social) norms. If this is right, society
shirking resulted in declining worker effort. Fehr and may also influence which obligations or norms are
Falk (2002: 695) conclude that bin the presence of governing a certain area. Norms may dcrowd inT.
non-pecuniary motives, there are important and, Nyborg (2003) gives an illustration of this in the case
relative to the predictions of the economic model, of smoking regulations in the public sphere which
unexpected interactions between material incentives resulted in changed behavior also in the private
and non-pecuniary motivesQ. They refer to several domain. The regulation implied a shift from smoking
studies, like Bohnet et al. (2001) and Schulze and being dfreeT to becoming something one had to ask for
Frank (2003), with similar results. also in domains outside where the law was actually
In yet another set of studies, so-called dcrowding functioning.
outT effects are observed. Most typically we have How can we understand the various observations
cases where paying for a good decreases supply or referred to above? The picture is rather complex, and
willingness to cooperate—(e.g., Frey, 1997a; Frey and there are variations not only between institutional
Oberholzer-Gee, 1997; Frey and Jegen, 2001). This is settings, but also between individuals in the same
observed in situations where payment is not found to setting. This implies that there is both an effect of the
be adequate since the supply is considered to be an institutional structure, formal or informal, and the
obligation. Blood donation is a classical example, propensity of various individuals to act more or less
where a shift from free donation to invoking payments selfishly. Putting emphasis to the institutional aspect,
has been observed to result in less supply (Titmus, Fig. 1 may help to make sense of much of what has
cited in Frey, 1997a). Thus, according to these been observed.
authors, problems appear when the incentive mecha- The figure is based on the synthesis made in
nism used does not conform well to the logic of the Section 2.3. It distinguishes between what we may
concrete situation as perceived by the respondents or call dIT and dWeT motivated behavior, between on the
suppliers. Parallel to this is the situation where the one hand selfish or individually strategic acting and
institutional setting influences more directly which on the other hand reciprocal behavior or behavior
preferences are evoked. Romer (1996)13 documents based on norms/obligations. The figure focuses on the
how income transfer programs shape preferences. principal–agent relation, but the reasoning is relevant
also for the results observed in ultimatum games/
13
I am indebted to Bowles (1998) for this reference. public goods games.
214 A. Vatn / Ecological Economics 55 (2005) 203–217

In quadrants A&D the principal and agent per- conditions characterized by pure strategic behavior
ceives the rationality or logic of the situation in an faces the danger of becoming unstable since they tend
equal manner. In A they see it as governed by to weaken their own fundament, the level of trust.
selfishness/strategic action. In D they see it as ruled by This is so, since trust requires a certain level of
reciprocity or obligations. In the first case economic common norms (a dWeT) to persist. The effect will be
incentives function well. In the latter voluntary increased transaction costs implying more detailed
compliance is the rule. contracts, more control, etc. Many recent cases from
Quadrant B is typically describing a dcrowding outT the business sector illustrate this development.
situation. It may, however, also apply to situations The above reasoning illustrates how policy instru-
concerning behavioral controls and incentive wages. ments may affect motivation structures. It emphasizes
The principal believes that the agent is acting the potential high costs of using the wrong instru-
strategically, while the agents look at the situation as ments. While reciprocity and commitments on behalf
being one of obligations or reciprocity. They find the of the agents may easily ddeteriorateT into strategic
situation governed by norms, trust and/or commit- behavior if the incentive schemes assume such
ments. If the principal uses economic incentives and behavior, building trust, reciprocity and commitment
expects a calculative response, s/he will observe seems to be a much more complex and difficult
dperversionsT. People are offered money for some- process. Thus there is an asymmetry between boxes
thing considered a duty. Dependent on the situation, B&C. What is easy to tear down, often cannot be
the use of a selfregarding type of instrument may easily built up again. Trust cannot be established just
make people shift motivation, i.e., accept that the by acting as if agents behave reciprocally. Instead,
situation is now to be governed by policies directed at existing commitments may deteriorate under the
individual gain. We are moved to quadrant A. influence of incentives where those playing strategi-
Quadrant C describes a situation where the cally will systematically win. This happens if the
principal formulates a normative policy while agents principal chooses a dWeT strategy and not all agents
act strategically. In this case the principal is vulnerable act according to the same rule. As a consequence of
to being cheated by the agents. It may, however, also this, we see that also the D quadrant is vulnerable to
induce shifts in motivations as far as the norm deterioration, here in the form of free riding (a move
becomes internalized. The example with smoking to C). Those acting strategically in a situation
regulations illustrates that this may happen rather dominated by committed agents will gain in the form
smoothly, at least in cases where some agents gain of increased individual pay-offs. Controlling these
directly from such a shift in norms. agents will tend to increase the commitment of others
An interesting issue concerns the dynamics of each to the regulation.
quadrant. A&D depict stable situations. In the case of
B one might fear a deterioration of the dsocial capitalT
embedded in the norms of reciprocity or the commit- 4. Conclusion
ments involved. A situation characterized by cooper-
ative will deteriorates into a situation of pure strategic This paper has focused on the differences between
behavior where a great deal of control and direct the individualist and the social constructivist positions
punishment are needed. Etzioni (1988) emphasizes concerning behavior and institutions. We have
this as a general problem in modern societies. He goes observed that there is a strong relationship between
one step further by underlining the danger of this the model of behavior and the definition of institutions
process in the end undermining markets themselves, used. Neoclassical economic theory is individualist. It
since they also depend on some level of trust. Bowles looks at the individual as an instrumentally rational
(1998) draws similar conclusions. entity, self-contained entity having given preferences.
Actually, these observations raise the question As far as institutions are attached to this model of the
about the stability of quadrant A, which was taken human, they are seen as external rules established as
for granted above. Since transaction costs are positive, constraints on individual behavior to simplify coordi-
trust is necessary in any contracting situation. Thus nation and reduce transaction costs.
 A. Vatn / Ecological Economics 55 (2005) 203–217 215

According to the model of social construction, the Furthermore, there is growing evidence that the
individual acts on the basis of cognitive and norma- policy instruments used, again through the type of
tive structures that are socially constructed. Institu- logic they invoke, also influence preferences and
tions are not only constraints. They are what shape the behavior. People apply different behaviors in different
individual and define which rationality is relevant or institutional settings. This neatly fits the cognitive
appropriate in each type of setting. model where these issues are two sides of the same
Both positions focus very much on the issue of coin. Choosing policy instruments is thus not simply
creating order. In the constructivist model the problem about changing incentives. First of all it is about
of order is one about creating common frames of instituting certain logics, about understanding which
reference, common typifications and norms. The institutional frames people apply, and about influenc-
neoclassical economic model does not address these ing these frames.
issues. The problem of order is here that of structuring These observations raise a set of issues relevant
exchanges between given individuals in a non- for the analysis of environmental questions. The
communicative way. most basic challenge is to choose which institutional
The two models can be brought together on the frame best fits the issues at hand. This second order
basis of the cognitivist tradition. From this perspective question must be evaluated both in the short and
the economic model can be understood as a dspecial long run. There are two arguments for supporting a
caseT. It is a situation where the individual is more citizens and norm oriented focus. First,
supported by institutions that promote instrumental environmental issues are basically about the com-
rationality. This is one out of many institutional mon good, about how we interact in each other’s
structures under which people can behave and get lives. Second, constructing institutions that em-
socialized. Other settings are more rule - or norm- phasize a citizen’s perspective and the common
based. Yet, all of them are governed by the basic responsibilities involved supports the strengthening
mechanisms of social construction. of trust and engagement in forming the common
From this perspective important second order good that environmental policy in the end so
questions follow: How do various institutional settings crucially depends upon.
influence people? Under which institutional frame is a
specific issue best treated? In the end, these are
questions about who we wants to become, an issue
Acknowledgement
that is invisible when we use a model that takes the
individual as given and institutionally independent.
The author would like to thank Jouni Paavola and
In environmental policy these are fundamental
P3l Vedeld for comments to earlier drafts.
issues. First, social structures form individual per-
spectives and preferences. If it is observed that people
do not have preferences for some environmental
goods, this may be because issues are typically novel References
and social processes necessary to form perspectives
and preferences have not yet taken place. Second, if Akerlof, G.A., 1970. The market for dLemonsT: quality uncertainty
people have problems with perceiving environmental and the market mechanism. Quarterly Journal of Economics 84,
488 – 500.
issues in monetary terms, it may be because they Arrow, K., 1951. Social choice and individual values. Yale
invoke rationalities that differ from the market logic University Press, New Haven.
when thinking about environmental issues. For them Becker, G., 1976. The economic approach to human behavior.
invoking this market logic when dealing with the University of Chicago Press, Chicago.
environment may be a category mistake. This Berger, P., Luckman, T., 1967 [1991]. The Social Construction of
Reality. A Treatise in the Sociology of Knowledge. Penguin
supports the need for developing institutional arrange- Books, London.
ments that can better take account of the multi- Bohnet, I., Frey, B.S., Huck, S., 2001. More order with less law: on
dimensionality of values characterizing environmental contract enforcement, trust, and crowding. American Political
choices. Science Review 95 (1), 131 – 144.
216 A. Vatn / Ecological Economics 55 (2005) 203–217

Bowles, S., 1998. Endogenous preferences: the cultural consequen- Frey, B.S., Jegen, R., 2001. Motivation crowding theory. Journal of
ces of markets and other economic institutions. Journal of Economic Surveys 15 (5), 589 – 611.
Economic Literature XXXVI, 75 – 111 (March). Frey, B.S., Oberholzer-Gee, F., 1997. The cost of price incentives:
Bromley, D.W., 1989. Economic Interests and Institutions. The an empirical analysis of motivation crowding-out. American
Conceptual Foundations of Public Policy. Basil Blackwell, Economic Review 87 (4), 746 – 755.
Oxford. Fudenberg, D., Tirole, J., 1991. Game Theory. The MIT Press,
Buchanan, J., 1978. From private preferences to public philosophy. Cambridge, MA.
In: Buchanan, J. (Ed.), The Economics of Politics. Institute of Furubotn, E.G., Richter, R., 1998. Institutions and Economic
Economic Affairs, London. Theory. The Contribution of the New Institutional Economics.
Clark, J., Burgess, J., Harrison, C.M., 2000. bI struggled with this The University of Michigan Press, Michigan.
money businessQ: respondents’ perspectives on contingent Gintis, H., 2000. Beyond homo economicus: evidence from
valuation. Ecological Economics 33, 45 – 62. experimental economics. Ecological Economics 35, 311 – 322.
Coase, R.H., 1960. The problem of social cost. The Journal of Law Gneezy, U., Rustichini, A., 2000. Pay enough or don’t pay at all.
and Economics 3, 1 – 44. The Quarterly Journal of Economics 115 (2), 791 – 810.
Dahlman, C.J., 1979. The problem of externality. The Journal of Gowdy, J., 2004. The revolution in welfare economics and its
Law and Economics 22, 141 – 162. implications for environmental valuation and policy. Land
Dearlove, J., 1989. Neoclassical politics: public choice and political Economics 80 (2), 239 – 257.
understanding. Review of Political Economy 1, 208 – 237. Gregory, R., Lichtenstein, S., Slovic, P., 1993. Valuing environ-
Dequech, D., 2002. The demarcation between the bOldQ and the mental resources: a constructive approach. Journal of Risk and
QNewQ institutional economics: recent complications. Journal of Uncertainty 7, 177 – 197.
Economic Issues XXXVI (2), 565 – 572. Groenewegen, J., Pitelis, C., Sjfstrand, S.E., 1995. Introduction. In:
Diamond, P.A., Hausman, J.H., 1994. Contingent valuation: is some Groenewegen, J., Pitelis, C., Sjfstrand, S.E. (Eds.), On
number better than no number? Journal of Economic Perspec- Economic Institutions—Theory and Applications. Edward
tives 8 (4), 45 – 64. Elgar, Cheltenham, pp. 1 – 6.
Douglas, M., 1986. How Institutions Think. Syracuse University Güth, W., Schmittberger, R., Schwarze, B., 1982. An experimental
Press, Syracuse. analysis of ultimatum bargaining. Journal of Economic Behav-
Durkheim, E., 1893 [1949]. The division of labor in society. Free ior and Organization 3, 367 – 388.
Press, Glencoe, IL. Habermas, J., 1984. The Theory of Communicative Action. Beacon
Durkheim, E., 1895 [1950]. The Rules of Sociological Method. Free Press, Boston.
Press, Glencoe, IL. Hanemann, W.M., 1994. Valuing the environment through
Eggertsson, T., 1990. Economic Behavior and Institutions. Cam- contingent valuation. Journal of Economic Perspectives 8 (4),
bridge Survey of Economic Literature, Cambridge. 19 – 43.
Elster, J., 1983. Explaining Technical Change: A Case Study in Hausman, D.M., 1992. The Inexact and Separate Science of
the Philosophy of Science. Cambridge University Press, Economics. Cambridge University Press, Cambridge.
Cambridge. Hodgson, G.M., 1998. The approach of institutional economics.
Elster, J., 1990. The Commentators. Jon Elster. In: Swedberg, R. Journal of Economic Literature XXXVI, 166 – 192. (March).
(Ed.), Economics and Sociology. Redefining their Boundaries: Holland, A., 1995. The assumptions of cost–benefit analysis: a
Conversations with Economists and Sociologists. Princeton philosopher’s view. In: Willis, K.G., Corkindale, J.T. (Eds.),
University Press, Princeton, pp. 233 – 248. Environmental Valuation. New Perspectives. CAB International,
Etzioni, A., 1988. The Moral Dimension: Toward a New Wallingford, pp. 21 – 38.
Economics. The Free Press, New York. Jacobs, M., 1997. Environmental valuation, deliberative democracy
Fehr, E., Falk, A., 2002. Psychological foundations of incentives. and public decision-making institutions. In: Foster, J. (Ed.),
European Economic Review 46, 687 – 724. Valuing Nature? Economics, Ethics and Environment. Rout-
Fehr, E., Gächter, S., 2000. Co-operation and punishment in ledge, London, pp. 211 – 231.
public goods experiments. American Economic Review 90 (4), Kapp, K.W., 1971. The Social Costs of Private Enterprise. Schoken
980 – 994. Books, New York.
Fehr, E., Gächter, S., 2002. Do incentive contracts undermine Knudsen, C., 1993. Equilibrium, perfect rationality and the problem
voluntary co-operation? Working paper no 34, Institute for of self-reference in economics. In: Mäki, U., Gustafsson, B.,
Empirical Research in Economics, University of Zürich. Knudsen, C. (Eds.), Rationality, Institutions and bEconomic
Fehr, E., Gächter, S., Kirschsteiger, G., 1997. Reciprocity as a MethodologyQ. Routledge, London, pp. 133 – 170.
contract enforcement device: experimental evidence. Econo- Lane, R.E., 1991. The Market Experience. Cambridge University
metrica 65 (4), 833 – 860. Press, Cambridge.
Frey, B.S., 1997a. Not Just for the Money. An Economic March, J.G., Olsen, J.P., 1989. Rediscovering Institutions. The
Theory of Personal Motivation. Edward Elgar Publishing, Organizational Basis of Politics. Free Press, New York.
Cheltenham. Munda, G., 2004. Social multi-criteria evaluation: methodological
Frey, B.S., 1997b. A constitution for knaves crowds out civic virtue. foundations and operational consequences. European Journal of
Economic Journal 107 (443), 1043 – 1053. Operational Research 158 (3), 662 – 677.
 A. Vatn / Ecological Economics 55 (2005) 203–217 217

Niskanen, W., 1971. Bureaucracy and Representative Government. Spash, C.L., 2000. Multiple value expression in contingent
Aldine-Atherton, Chicago. valuation: economics and ethics. Environmental Science and
Nbrretranders, T., 1991. M&rk V&rden. En beretning om bevidsthet. Technology 34, 1433 – 1438.
Gyldendal, Kbbenhavn. Spash, C.L., Hanley, N., 1995. Preferences, information and
North, D.C., 1990. Institutions, Institutional Change and Economic biodiversity preservation. Ecological Economics 12, 191 – 208.
Performance. Cambridge University Press, Cambridge. Stevens, T.H., Echeverria, J., Glass, R.J., Hager, T., Moore, T.A.,
Nyborg, K., 2003. The impact of public policy on social and 1991. Measuring the existence value of wildlife: what do CVM
moral norms: some examples. Journal of Consumer Policy 26, estimates really show. Land Economics 67, 390 – 400.
258 – 277. Sunstein, C.R., 1993. Endogenous preferences, environmental law.
O’Neill, J., 1993. Ecology, Policy and Politics. Human Well-being Journal of Legal Studies 22 (2), 217 – 254.
and the Natural World. Routledge, London. Tenbrunsel, A.E., Messick, D.M., 1999. Sanctioning systems,
Ostrom, E., 2000. Collective action and the evolution of social decision frames and co-operation. Administrative Science
norms. Journal of Economic Perspectives 14 (3), 137 – 158. Quarterly 44, 684 – 707.
Romer, P., 1996. Preferences, promises and the politics of Tool, M.R., 1995. Pricing, Valuation and Systems. Essays in
entitlements. In: Fuchs, V. (Ed.), Individual and Social Neoinstitutional Economics. Edward Elgar, Aldershot.
Responsibility: Child Care, Education, Medical Care and Vandjal, D., O’Connor, M., 1994. What is the value of Rangitoto
Long-term Care in America. University of Chicago Press, Island? Environmental Values 3, 369 – 390.
Chicago, pp. 195 – 228. Vatn, A., 2004. Environmental valuation and rationality. Land
Romp, G., 1997. Game Theory. Introduction and Application. Economics 80 (1), 1 – 18.
Oxford University Press, Oxford. Vatn, A., 2005. Institutions and the Environment. Edward Elgar,
Rousseau, J.-J., 1762 [1968]. The Social Contract. Penguin, Cheltenham.
Harmondsworth. Veblen, T., 1898. Why is economics not an evolutionary science.
Schkade, D.A, Payne, J.W., 1993. Where do the numbers Quarterly Journal of Economics 12, 373 – 397.
come from? How people respond to contingent Veblen, T., 1899. The Theory of the Leisure Class: An Economic
valuation questions. In: Hausman, J.A. (Ed.), Contingent Study of Institutions. Macmillan, New York.
Valuation. A Critical Assessment. North Holland, Amsterdam, Walzer, M., 1983. Spheres of Justice. A Defence of Pluralism and
pp. 271 – 293. Equality, Basic Books, Publishers, New York.
Schulze, G.G., Frank, B., 2003. Deterrence versus intrinsic Williamson, O.E., 1985. The Economic Institutions of Capitalism.
motivation: experimental evidence on the determinants of Free Press, New York.
corruptibility. Economics of Governance 4, 143 – 160. Williamson, O.E., 2000. The new institutional economics: taking
Scott, W.R., 1995. Institutions and Organizations. Sage Publica- stock/looking ahead. Journal of Economic Literature XXXVIII
tions, California. (3), 595 – 613.